Skip to main content
ORNL DAAC HomeNASA Home

DAAC Home > Get Data > Regional/Global > Vegetation Collection > Data Set Documentation
Long-Term Arctic Growing Season NDVI Trends from GIMMS 3g, 1982-2012
Get Data

Revision Date: March 11, 2015

Summary

This data set provides normalized difference vegetation index (NDVI) data for the arctic growing season derived primarily with data from Advanced Very High Resolution Radiometer (AVHRR) sensors onboard several NOAA satellites over the years 1982 through 2012. The NDVI data, which show vegetation activity, were averaged annually for the arctic growing season (GS; June, July and August). The products include the annual GS-NDVI values and the results of a cumulative GS-NDVI time series trends analysis. The data are circumpolar in coverage at 8-km resolution and limited to greater than 20 degrees N.

These normalized difference vegetation index (NDVI) trends were calculated using the third generation data set from the Global Inventory Modeling and Mapping Studies (GIMMS 3g). GIMMS 3g improves on its predecessor (GIMMS g) in three important ways. First, GIMMS 3g integrates data from NOAA-17 and 18 satellites to lengthen its record. Second, it addresses the spatial discontinuity north of 72 degrees N, by using SeaWiFS, in addition to SPOT VGT, to calibrate between the second and third versions of the AVHRR sensor (AVHRR/2 and AVHRR/3). Finally, the GIMMS 3g algorithm incorporates improved snowmelt detection and is calibrated based on data from the shorter, arctic growing season (May-September) rather than the entire year (January-December).

The annual GS-NDVI data are provided in a single NetCDF v4 file. The GS-NDVI trends, significance of these trends, and a land cover map (GLC2000) are provided as three GeoTIFF files.

NDVI Arctic Growing Season Trends from GIMMS 3G, 1982-2012


Figure 1: NDVI Arctic Growing Season Trends from GIMMS 3g, 1982-2012.

ACKNOWLEDGEMENTS

We acknowledge support from the NASA Carbon Cycle and Ecosystems program grants NNX08AG13G and NNX09AI24G to S.J. Goetz.

Data and Documentation Access:

Get Data: http://daac.ornl.gov/cgi-bin/dsviewer.pl?ds_id=1275

Data Citation:

Cite this data set as follows:

Guay, K.C., P.S.A. Beck, and S.J. Goetz. 2015. Long-Term Arctic Growing Season NDVI Trends from GIMMS 3g, 1982-2012 . Data set. Available on-line [http://daac.ornl.gov] from Oak Ridge National Laboratory Distributed Active Archive Center, Oak Ridge, Tennessee, USA. http://dx.doi.org/10.3334/ORNLDAAC/1275

Table of Contents:

1. Data Set Overview:

This data set provides normalized difference vegetation index (NDVI) data for the arctic growing season derived primarily with data from the Advanced Very High Resolution Radiometer (AVHRR) aboard several NOAA satellites over the years 1982 through 2012. The NDVI data were averaged annually for the arctic growing season (GS; June, July and August). The products include the annual GS-NDVI values and the results of a cumulative GS-NDVI time series trends analysis. The data are circumpolar in coverage at 8-km resolution and limited to greater than 20 degrees N.

This third generation Global Inventory Modeling and Mapping Studies (GIMMS 3g) product is a ca. 8-km resolution, 15-day maximum value composite (MVC), bimonthly, global NDVI product generated from AVHRR data (NOAA-7, 9, 11, 14, 16-19). GIMMS 3g integrates data from NOAA-17 and -18, to improve the length and quality of the GIMMS-NDVI record and uses SeaWiFS (along with SPOT VGT) data to combine the AVHRR/2 and AVHRR/3 data sets to address the discontinuity north of 72 degrees N, present in GIMMSg (Pinzon et al., 2007; Pinzon and Tucker 2014). The GIMMS 3g algorithm also has improved snow-melt detection and is calibrated based on data from the shorter, arctic growing season (May–September), rather than the entire year (January–December).

GS-NDVI time series trends (brownness-greenness) were calculated using the Theil-Sen method (Theil, 1992). Areas defined as cropland, permanent snow and ice or other unnatural vegetation types in the Global Land Cover 2000 map (GLC2000; Table 1) were masked out. The GLC2000 map has been resampled (using the nearest neighbor method) to match the GIMMS 3G 8-km grid and is included in this data set (Guay et al. 2014).

2. Data Characteristics:

Spatial Coverage

Northern hemisphere north of 20 degrees N.

Spatial Resolution

8 km

Site boundaries: (All latitude and longitude given in decimal degrees)

Westernmost Longitude Easternmost Longitude Northernmost Latitude Southernmost Latitude
-18018090 20

 

Temporal Coverage

AVHRR source data from 1982-06-01 to 2012-08-31.

Temporal Resolution

Annual growing season average (June, July, August) and 31-year compiled trend results

Data File Information

The annual GS-NDVI data are proved in a single NetCDF v4 file. The GS-NDVI trends, the significance of trends, and the land cover map (GLC2000) are provided in GeoTIFF format.

GS-NDVI Trends

gimms3g_gs_ndvi_trend_8km_1982_2012_20deg_trend.tif

gimms3g_gs_ndvi_trend_8km_1982_2012_20deg_significance.tif

gimms3g_gs_ndvi_trend_8km_1982_2012_20deg_landcover.tif

GeoTIFF Spatial Data Properties

Spatial Representation Type: Raster
  Compression Type: LZW
  Number of Bands: 1
  Raster Format: GeoTIFF
  No Data Value: -1.79769313486232e+308
  Scale Factor: 1

Number Columns: 4320
  Column Resolution: 8 km
  Number Rows: 840
  Row Resolution: 8 km

Extent in the items coordinate system
  North: 90
  South: 20
  West: -180
  East: 180

Spatial Reference Properties

Type: Geographic
 Geographic Coordinate Reference: WGS 1984

Open Geospatial Consortium (OGC) Well Known Text (WKT)
  GEOGCS["WGS 84",
  DATUM["WGS_1984",
  SPHEROID["WGS 84",6378137,298.257223563,
   AUTHORITY["EPSG","7030"]],
   AUTHORITY["EPSG","6326"]],
  PRIMEM["Greenwich",0],
  UNIT["degree",0.0174532925199433],
   AUTHORITY["EPSG","4326"]]
Origin = (-180.000000000000000,90.000000000000000)
Pixel Size = (0.083333333333333,-0.083333333333333)

Land Cover Classes

Global Land Cover 2000 Map (GLC2000) was resampled to match the GIMMS 3g grid. The following classes were masked in Guay et al. (2014).

VALUE CLASS NAME MASKED
1 Tree Cover, broadleaved, evergreen NO
2 Tree Cover, broadleaved, deciduous, closed NO
3 Tree Cover, broadleaved, deciduous, open NO
4 Tree Cover, needle-leaved, evergreen NO
5 Tree Cover, needle-leaved, deciduous NO
6 Tree Cover, mixed leaf type NO
7 Tree Cover, regularly flooded, fresh water NO
8 Tree Cover, regularly flooded, saline water NO
9 Mosaic: Tree Cover / Other natural vegetation NO
10 Tree Cover, burnt NO
11 Shrub Cover, closed-open, evergreen NO
12 Shrub Cover, closed-open, deciduous NO
13 Herbaceous Cover, closed-open NO
14 Sparse herbaceous or sparse shrub cover NO
15 Regularly flooded shrub and/or herbaceous cover NO
16 Cultivated and managed areas YES
17 Mosaic: Cropland / Tree Cover / Other natural vegetation YES
18 Mosaic: Cropland / Shrub and/or grass cover YES
19 Bare Areas YES
20 Water Bodies YES
21 Snow and Ice YES
22 Artificial surfaces and associated areas YES
23 No data YES

Annual GS-NDVI data in NetCDF v4 file

gimms3g_ndvi_1982-2012.nc4

NetCDF v4 Header -- CF compliant

Parameter, Parameter Abbreviation, Unit, and Description

  Parameter: GIMMS 3G NDVI in growing season (June, July, and August)
   Parameter Abbreviation: NDVI
Unit: GS-NDVI
Description: Mean Normalized Difference Vegetation Index in growing season (June, July, and August)

Spatial Data Properties

  Spatial Representation Type: Raster
  Pixel Depth: 64 bit
   Pixel Type: double precision
  Number of Bands: 31
   Band Information: time
  Raster Format: netCDF
  Source Type: Generic
  No Data Value: -9999
  Scale Factor: none
  Number Columns: 4320

Column Resolution: 8 km
  Number Rows: 840
Row Resolution: 8 km

Extent in the items coordinate system
  North: 90
  South: 20
  West: -180
  East: 180

Spatial Reference Properties

Type: Geographic
  Geographic Coordinate Reference: WGS 1984

Open Geospatial Consortium (OGC) Well Known Text (WKT)
  GEOGCS["WGS 84",
   DATUM["WGS_1984",
    SPHEROID["WGS 84",6378137,298.257223563,
  AUTHORITY["EPSG","7030"]],
  AUTHORITY["EPSG","6326"]],
  PRIMEM["Greenwich",0],
  UNIT["degree",0.0174532925199433],
   AUTHORITY["EPSG","4326"]]
  Origin = (-180.000000000000000,90.000000000000000)
  Pixel Size = (0.083333333333333,-0.083333333333333)

Data Archive Center Processing:

The single netCDF file was created by combining 31 individual annual GS-NDVI GeoTIFFs using Geospatial Data Abstraction Library (GDAL) and netCDF Operator (NCO) software. The NCO software was then utilized to edit the netCDF header to bring it into Climate and Forecast (CF) compliance.

3. Data Application and Derivation:

The AVHRR-based GIMMS-NDVI version G (GIMMS g) series, and its recent successor version 3g (GIMMS 3g), as well as the shorter NDVI records generated from the more modern sensors, SeaWiFS, SPOT-VGT, and MODIS were compared. The data sets from the latter two sensors were provided in a form that reduces the effects of surface reflectance associated with solar and view angles (Guay et al., 2014).

4. Quality Assessment:

See Pinzon and Tucker (2014) for discussion of specific comparisons between GIMMS 3g and independently measured appropriate climate phenomena needed to improve our understanding of uncertainty the AVHRR GIMMS 3g product.

5. Data Acquisition Materials and Methods

This data set provides normalized difference vegetation index (NDVI) data for the arctic growing season derived primarily with data from Advanced Very High Resolution Radiometer (AVHRR) sensors aboard several NOAA satellites over the years 1982 through 2012. The NDVI data were averaged annually for the arctic growing season (GS; June, July and August).

The data are circumpolar in coverage at 8-km resolution and limited to greater than 20 degrees N. Like its predecessor, the third generation Global Inventory Modeling and Mapping Studies (GIMMS 3g) product is a ca. 8-km resolution, 15-day maximum value composite (MVC), bimonthly, global NDVI product generated from AVHRR data (NOAA-7, 9, 11, 14, 16-19). GIMMS 3g integrates data from NOAA-17 and -18, to improve the length and quality of the GIMMS-NDVI record and uses SeaWiFS (along with SPOT VGT) data to combine the AVHRR/2 and AVHRR/3 data sets to address the discontinuity north of 72 degrees N, present in GIMMS g (Pinzon et al., 2007; Pinzon and Tucker 2014). The GIMMS 3g algorithm also has improved snowmelt detection and is calibrated based on data from the shorter, arctic growing season (May–September), rather than the entire year (January–December).

As described in Guay et al. (2014), the temporal trends in the GS-NDVI annual time series were estimated using the Theil-Sen approach (Theil, 1992). Trends were not estimated for pixels that were missing data anywhere in the time series. The Mann-Kendall test was used to assess trends for statistical significance (Mann, 1945), regarding pixels as significant when P < 0.05.

Compared to a simple linear regression, the Theil-Sen slope estimator is more robust to outliers, as it estimates the slope of a time series as the median of all slopes between pairs of observations in the time series. We explored using this population of slopes to empirically determine 95% confidence intervals around the slope estimate as an alternative assessment of statistical significance. However, the Mann-Kendall test was far more conservative when applied to the common record. Trend analyses were performed using the tools in the zyp package (Bronaugh & Werner, 2012) in R (R Core Team, 2012).

The GIMMS 3g data product used as a source for this data set and those described in Guay et al. (2014) in Pinzon and Tucker (2014), are available from the NASA Earth Exchange (NEX), a platform for scientific collaboration, knowledge sharing and research for the earth science community. Access link: https://nex.nasa.gov/nex/projects/1349/

6. Data Access:

These data are available through the Oak Ridge National Laboratory (ORNL) Distributed Active Archive Center (DAAC).

Data Archive Center:

Contact for Data Center Access Information:

E-mail: uso@daac.ornl.gov
Telephone: +1 (865) 241-3952

7. References:

Bronaugh D, Werner A (2012) zyp: Zhang + Yue-Pilon Trends Package. R Foundation for Statistical Computing, Vienna.

Guay KC, Beck PSA, Berner LT, Goetz SJ, Baccini A, Buermann W (2014) Vegetation productivity patterns at high northern latitudes: a multi-sensor satellite data assessment. Global Change Biology, 20, 3147-3158.

Mann HB (1945) Nonparametric tests against trend. Econometrica, 13, 245-259.

Pinzon JE, Brown ME, Tucker CJ (2007) Global Inventory Modeling and Mapping Studies (GIMMS) Satellite Drift Corrected and NOAA-16 incorporated Normalized Difference Vegetation Index (NDVI), Monthly 1981-2006.

Pinzon JE, Tucker CJ (2014) A Non-Stationary 1981-2012 AVHRR NDVI3g Time Series. Remote Sensing, 6, 6929-6960.

R Core Team (2012) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.

The Global Land Cover Map for the Year 2000, 2003. GLC2000 database, European Commission Joint Research Centre. http://www-gem.jrc.it/glc2000.

Theil H (1992) A rank-invariant method of linear and polynomial regression analysis. In: Econometric Theory and Methodology, Vol. 23 (eds Raj B, Koerts J), pp. 345–381. Springer, Houten, the Netherlands.